The Woody-Preferential Gene EgMYB88 Regulates the Biosynthesis of Phenylpropanoid-Derived Compounds in Wood

Abstract

Comparative phylogenetic analyses of the R2R3-MYB transcription factor family revealed that five subgroups were preferentially found in woody species and were totally absent from Brassicaceae and monocots (Soler et al., 2015). Here, we analyzed one of these subgroups (WPS-I) for which no gene had been yet characterized. Most Eucalyptus members of WPS-I are preferentially expressed in the vascular cambium, the secondary meristem responsible for tree radial growth. We focused on EgMYB88, which is the most specifically and highly expressed in vascular tissues, and showed that it behaves as a transcriptional activator in yeast. Then, we functionally characterized EgMYB88 in both transgenic Arabidopsis and poplar plants overexpressing either the native or the dominant repression form (fused to the Ethylene-responsive element binding factor-associated Amphiphilic Repression motif, EAR). The transgenic Arabidopsis lines had no phenotype whereas the poplar lines overexpressing EgMYB88 exhibited a substantial increase in the levels of the flavonoid catechin and of some salicinoid phenolic glycosides (salicortin, salireposide, and tremulacin), in agreement with the increase of the transcript levels of landmark biosynthetic genes. A change in the lignin structure (increase in the syringyl vs. guaiacyl, S/G ratio) was also observed. Poplar lines overexpressing the EgMYB88 dominant repression form did not show a strict opposite phenotype. The level of catechin was reduced, but the levels of the salicinoid phenolic glycosides and the S/G ratio remained unchanged. In addition, they showed a reduction in soluble oligolignols containing sinapyl p-hydroxybenzoate accompanied by a mild reduction of the insoluble lignin content. Altogether, these results suggest that EgMYB88, and more largely members of the WPS-I group, could control in cambium and in the first layers of differentiating xylem the biosynthesis of some phenylpropanoid-derived secondary metabolites including lignin.

Keywords: Eucalyptus; MYB transcription factors; flavonoids; lignin; oligolignols; phenylpropanoid metabolism; salicinoid phenolic glycosides; vascular cambium.

Figure 1

Neighbor-Joining phylogenetic tree constructed with R2R3-MYB members of phenylpropanoid metabolism-related super clade. The proteins included belong to the subgroups S4, S5, S6, S7, S15, SAtMYB5, SAtMYB82, WPS-I, WPS-II, and WPS-III from E. grandis (red dots), V. vinifera (pink dots), P. trichocarpa (yellow dots), and A. thaliana (blue dots). The distant protein AtMYB52 (Cassan-Wang et al., 2013), involved in the lignin biosynthesis, was used to root the tree. The WPS-I is expanded to show the different members from E. grandis, P. trichocarpa and V. vinifera, and the absence of genes from Arabidopsis. EgrMYB88 is highlighted with a gray background. Bootstrap values are shown on the internodes.

Figure 2

Amino acid sequence alignment of the E. grandis and P. trichocarpa genes belonging to the WPS-I. Amino acid are colored according to their similarity degree using the ClustalX option of Jalview (Waterhouse et al., ; http://www.jalview.org/help/html/colourSchemes/clustal.html). Light gray and dark gray rectangles at the bottom of the alignment indicate the R2 and the R3 MYB domains, respectively. Yellow rectangle indicates the bHLH interaction motif described by Zimmermann et al. (2004). Green rectangle indicates the region that, at the mRNA level, is targeted by the miR828 (Xia et al., 2012). Red rectangle indicates the position of the conserved C-terminal motif, whose sequence obtained using the MEME Suite (Bailey et al., 2009) is indicated below.

Figure 3

Sequence region targeted by miR828 among the E. grandis and P. trichocarpa members of WPS-I. (A) Gene nucleotide sequence alignment colored according to their degree of identity percentage using Jalview (Waterhouse et al., ; http://www.jalview.org/help/html/colourSchemes/pid.html). (B) Alignment between the EgMYB88 mRNA and the miR828 showing one single mismatch among 22 nucleotides. Sequence targeted by miR828 is based on Xia et al. (2012).

Figure 4

Transcript abundance data of the different Eucalyptus genes from WPS-I. Expression data (means ± standard deviation) in different organs and tissues from Eucalyptus was extracted from Soler et al. (2015). Vascular tissues are highlighted with a gray background. EgrMYB88 is the gene with the highest expression in vascular tissues.

Figure 5

Yeast auto-activation tests of EgMYB88 fused to the DNA binding domain of Gal4. (A) Reporter and effector constructs used in the assay. The reporter constructs consists of the cis-regulatory elements to which Gal4 binds before the four reporter genes tested. The effector constructs consist of fusion proteins with the Gal4 binding domain (BD) at the N-terminal end. As controls, we tested the empty vector as well as two well-known Eucalyptus genes, one acting as an activator (EgMYB2, Goicoechea et al., 2005) and the other as a repressor (EgMYB1, Legay et al., 2007). (B) Growth of yeast cells in different selective media. Growth in auxotrophic media (SD) without Tryptophan (Trp) indicates the presence of the effector plasmid. Growth in the absence of Histidine (His) or Adenine (Ade) are indicative of activation of HIS3 and ADE2 reporter genes. Growth in presence of the antibiotic Aureobasidin A (AurA) indicates the activation of AUR1-C, which confers resistance to this antibiotic. Blue color represent the activation of MEL1 reporter gene, able to produce a blue precipitate in presence of 5-bromo-4-chloro-3-indolyl alpha-D-galactopyranoside (X-α-Gal).

Figure 6

Transverse sections of inflorescence stems of transgenic Arabidopsis plants overexpressing EgMYB88 either as a native form or fused to an EAR motif. Detail of Arabidopsis inflorescence stems where lignin are stained in red by phloroglucinol-HCl. Scale bars: 50 μm. Images shown are representative of the three independent lines analyzed for each construction (Pro35S:EgMYB88 and Pro35S:EgMYB88-EAR, specified in Figure S1).

Figure 7

Transverse sections of stems from transgenic poplar plants overexpressing EgMYB88 either as a native form or fused to an EAR motif. (A) Detail of the cambial region and of the first layers of differentiating xylem cells. (B) Detail of mature xylem cells. Sections were stained using phloroglucinol-HCl which stains the lignin in red. Scale bars: 25 μm. Images shown are representative of the three independent lines analyzed for each construction (Pro35S:EgMYB88 and Pro35S:EgMYB88-EAR, specified in Figure S2). One single control representative of the two poplar batches is shown.

Figure 8

Transcript abundance of genes involved in the phenylpropanoid metabolism in transgenic poplars overexpressing EgMYB88 either as a native form or fused to an EAR motif. The genes are placed in a schematic representation of the pathways leading to lignin, flavonoids, and salicinoid phenolic glycosides, where relevant intermediates are shown (adapted from Takos et al., ; Peng et al., ; Jaakola, ; Carocha et al., ; Chedgy et al., ; Wang et al., 2015). Transcript abundance was expressed as a ratio of the abundance of a given gene in transgenic poplar plants, either overexpressing EgMYB88 (OE, shown at the left of the heatmap) or EgMYB88-EAR (EAR, shown at the right of the heat map), respective to its abundance in the corresponding controls. Values were calculated as the means of nine Pro35S:EgMYB88 (i.e., three plants for each of the three selected independent transgenic lines, specified in Figure S2A) and nine respective control poplar plants, or eight Pro35S:EgMYB88-EAR (i.e., two-three plants for each of the three selected independent transgenic lines, specified in Figure S2B) and seven respective control poplar plants. Statistical significance was calculated with Student's t-test, ^***P-value < 0.001, ^**P-value < 0.01, ^*P-value < 0.05. Gene names, expression data, and orthologs in other species are detailed in Tables 6, 7, Table S1.